Laserfiche WebLink
800 <br />w <br />700 <br />0 600 <br />w <br />500 <br />400 <br />300 <br />Ca <br />200 <br />100 <br />•N N <br />• • <br />• <br />0 100 200 300 400 . 500 600 700 800 <br />Cumulative ELOUI Seasonal Flow (in) <br />Figure 9: Cumulative discharge from Fool Creek (harvested) plotted over the cumulative discharge for <br />East St. Louis Creek (control) at the Fraser Experimental Forest. A represents the beginning <br />of measurement (1943), B represents the abrupt impact of harvest (1955), and C represents 1996. <br />As an example of the problems associated with defining the response from <br />timber harvest at the landscape scale, Kircher et al. (1985) evaluated those <br />factors most responsible for influencing natural stream flow characteristics <br />in western Colorado watersheds. Based on analysis of the long -term record <br />from 123 USGS gauging sites in western Colorado, only drainage area and <br />mean basin elevation were found to be significant parameters in predicting <br />mean annual flow. , Percentage of the total watershed area in forest, or <br />percentages of watershed area in other land use categories, were not a <br />significant parameter in predicting either annual mean flow or any other <br />stream flow characteristics Kircher et al. (1985) and others elsewhere have <br />been unable to show that the percentage of watershed in forest is correlated <br />with stream flow at landscape scale. In landscape scale watersheds, such as <br />those gauged by the USGS, forested area is often only a portion of the total <br />area. If the actual amount of forest vegetation is not well correlated with <br />flow at this level, change in the density of vegetation is equally unlikely to <br />be related as well. One can assume that detecting the effect of subtle <br />changes in forest vegetation over time, even where they occur, would be <br />difficult. As part of ongoing work (unpublished) by C. A. Troendle to <br />develop improved flood forecasting equations for western Colorado and <br />Wyoming, the percent of forested area in the watershed did not prove to be <br />significant in describing peak stream flow response at the landscape level; <br />i.e., using USGS stream flow data. At the landscape scale, many factors <br />influence stream flow generation, including the variance and errors <br />associated with monitoring flow at this scale and resolution; all of which <br />18 <br />f <br />1 <br />1 <br />